-
JOURNAL OF BACTERIOLOGY, Feb. 1972, p. 668-677Copyright 0 1972
American Society for Microbiology
Vol. 109, No. 2Printed in U.S.A.
Effect of Glycerol Deprivation on thePhospholipid Metabolism of
a GlycerolAuxotroph of Staphylococcus aureus
PAUL H. RAY AND DAVID C. WHITE
Biochemistry Department, University of Kentucky Medical School,
Lexington, Kentucky 40506
Received for publication 17 September 1971
A study of the effects of glycerol deprivation on the content
and metabolismof the phospholipids of a glycerol auxotroph of
Staphylococcus aureus showedthat (i) there was an increase in the
proportions of lysylphosphatidylglycerol(LPB) and a concomitant
decrease in the proportion of phosphatidylglycerol.The total
phospholipid content per sample and the proportion of
cardiolipindid not change, but the phosphatidic acid increased
transiently and then fell topretreatment levels. (ii) The loss Of
32P from the lipids during the chase in apulse-chase experiment was
essentially the same in phosphatidylglycerol, car-diolipin, and
phosphatidic acid during glycerol deprivation or growth in
thepresence of glycerol. LPG lost half the radioactivity in
slightly more than twodoubling times when grown with glycerol. In
the absence of glycerol, P2p accu-mulated in LPG for about 20 min
and then stopped, after which time there wasno apparent turnover.
(iii) During glycerol deprivation, the initial 32P incorpora-tion
decreased sixfold compared to that of the control with glycerol.
The initialincorporation into LPG decreased only 2.5-fold, whereas
that of PG decreased45-fold. (iv) During glycerol deprivation, the
free fatty acid content increasedfrom 1.2 to 12.5% of the total
extractable fatty acids and then slowly decreased.The increase was
largely iso- and anti-iso-branched 21-carbon-atom fatty acids.In
glycerol-supplemented cultures, the major fatty acids were branched
14- to18-carbon fatty acids. The decrease in longer chain free
fatty acids after 60 minrepresented their esterification into
lipids. (v) During glycerol deprivation ribo-nucleic acid synthesis
and cell growth continued for 40 min and protein syn-thesis
continued for 90 min. Then synthesis and growth stopped. (vi) After
theaddition of glycerol to glycerol-deprived cells, 32P and
14C-glycerol were incor-porated into the phospholipids without lag;
ribonucleic acid, protein synthesis,and cell growth began after a
5- to 10-min lag at the pretreatment rate. The ini-tial rate of
lipid synthesis after the addition of glycerol was three times
greaterthan the growth rate. This rapid rate continued for about 25
min until the lipidcontent and proportions of LPG and
phosphatidylglycerol were restored.
The involvement of phospholipids in thestructure and function of
the bacterial mem-brane has been the subject of intensive
investi-gation (18). Recent work has deepened theunderstanding of
how the bacterial phospho-lipids are involved in the formation and
func-tion of the membrane-wall complex. Some ofthe phospholiplds
have been shown to be es-sential in the function of the
lipopolysac-charide-forming enzymes (18) and in the ac-tivity of
enzyme II in the phosphotransferasefor the uptake of f3-glucosides
(8, 11). Cur-rently, mutants defective in the synthesis ofmembrane
components are being utilized to
correlate changes in the chemical and physicalproperties of the
membrane with changes intransport and in the activity of
membrane-bound enzymes (5, 14, 19). Mutants of Esche-richia coli
K-12 requiring unsaturated fattyacids were utilized to show the
requirement oflipid synthesis for the efficient induction ofvarious
transport proteins (3, 7). Others haveshown that by changing the
fatty acid compo-sition of the phospholipids the efficiency of
thelactose transport system can be modified (5,19; M. K. Crandell
et al., unpublished data).
Recently, Mindich has isolated glycerol aux-otrophs of both
Bacillus subtilis and Staphylo-
668
-
GLYCEROL AUXOTROPH OF S. AUREUS
coccus aureus (12-14). By using the B. subtilismutant, he has
shown that the citrate trans-port system can be induced under
conditionswhere de novo phospholipid synthesis isstopped although
ribonucleic acid (RNA) andprotein synthesis continue (27). However,
witha glycerol auxotroph of S. aureus, it was shownthat after
glycerol deprivation the lactosetransport system could be induced
and couldbe integrated into the membrane but did notfunction
efficiently. In further studies, Lillichand White (10) found that
even though netphospholipid synthesis stopped after
glyceroldeprivation there was considerable turnoverand resynthesis
of both the glycerol and phos-phate portions of the phospholipids
of B. sub-tilis.
In this study, a glycerol auxotroph of S. au-reus (S-2) obtained
from L. Mindich was uti-lized to investigate the metabolism of its
phos-pholipids so that future experiments could beconcerned with
the synthesis of electron trans-port system. Studies with
inhibitors have indi-cated that changes in the lipid
metabolismoccur concomitantly with the formation ofthe electron
transport system in S. aureus (9,25). In this paper we report the
effect of glyc-erol deprivation of the phospholipid composi-tion
and metabolism of a glycerol auxotroph ofS. aureus (S-2) under
conditions when proteinsynthesis continues.
MATERIALS AND METHODSGrowth of bacteria. The glycerol auxotroph
S-2
was derived from the parent strain S. aureus U-71 byL. Mindich
and was characterized as deficient innicotinamide adenine
dinucleotide-independent L-a-glycerol phosphate dehydrogenase
activity (15). Forall experiments, cells were grown in medium
con-taining 19 mm KCl, 0.49 mm K2HPO4, 79 mM NaCl,20 mm NH4C1, 1.4
mm Na2SO4, 0.1 mM adenine, 0.1mM xanthine, 0.1 mm uracil, 1.8 mM
alanine, 1.4 mMarginine, 0.75 mM asparagine, 1.5 mm cysteine,
0.75mM glutamic acid, 0.3 mm glycine, 1.1 mM histidine,2 mm
isoleucine, 4 mm leucine, 3.5 mM lysine, 1.1 mMmethionine, 0.09 mm
phenylalanine, 4.6 mm pro-line, 2.8 mM serine, 1.9 mM threonine,
0.08 mM tyro-sine, 2.5 mM valine, and 0.1 M
tris(hydroxymethyl)aminomethane. This medium was brought to pH
7.4with concentrated hydrochloride and autoclaved.After cooling,
the following solutions, which weresterilized either by filtration
or autoclaving, wereadded in the final concentrations given: 0.2 nM
bio-tin, 8.3 nM nicotinic acid, 0.93 nM thiamine, 2 gMFeCl8, 0.1 mM
CaCl2, 1.2 mM MgCl,, 0.1 mM tryp-tophan, 5.5 mM glucose, and 15 ug
of bovine serumalbumin (Sigma Chemical Co., St. Louis, Mo.) perml.
Cultures were grown in liter flasks containing400 ml of medium or
in Parrot flasks containing1,500 ml of medium in a Fermentation
Design Con-stant Temperature Water Bath Shaker (Allentown,
Pa.) at 37 C (0 1 C) and shaken at 200 rev/min.Growth was
measured as the absorbance at 750 nmwith a Spectronic 20
spectrophotometer. Dry weightsand lipid phosphate were determined
as previouslydescribed (17) and were linear with respect to
opticaldensity (Fig. 1).
Deprivation of glycerol. For studies requiringthe deprivation of
glycerol, cultures were grown toan absorbance between 0.2 and 0.4
in 500 ml of me-dium, harvested by rapid filtration on a
142-mmMillipore filter (0.4 gm pore size), washed with anequal
volume of prewarmed medium without glyc-erol, and resuspended in
warm medium with orwithout glycerol. There was no lag in the growth
ofthe bacteria after harvesting, washing, and resus-pending them in
the prewarmed medium.
Extraction of the lipids. The total lipids of S.aureus were
extracted by a modification of themethod of Bligh and Dyer (2) as
previously described(10). The cells in the growth medium were
immedi-ately added to chloroform and methanol to give aone-phase
system of chloroform-methanol-water (25:1:0.8). After extraction
for a minimum of 2 hr, thetwo-phase system was obtained by adding
chloro-form and methanol to give a final concentration
ofchloroform-methanol-water (1: 1: 0.9). The lipidphase was allowed
to separate and was dried by fil-tration over Na2SO4 before the
chloroform phase wasevaporated by flash evaporation.Chromatography
of the phospholipids. The
phospholipids of S. aureus were separated by two-dimensional
chromatography on silica gel-impreg-nated paper (Whatman SG-81) by
using the first andthird solvents of Wuthier (27). The first
dimension
JuMOLES LIPID PHOSPHATE/ ml (A).005 .010 .015 .020
0.8
EO0.7 00
0.6-
o0.4z49
0.30U,
0.Q2
02 0.3 0.4mg/ml DRY WEIGHT (0)
FIG. 1. Relationship between the absorbance at750 nm, the lipid
phosphate, and the dry weight ofStaphylococcus aureus S-2. Samples
from a growingculture were withdrawn and the absorbance at 750nm
was measured between 0.2 and 0.6 in a Spec-tronic 20
spectrophotometer. The sample was centri-fuged, washed, place in
preweighed glass vials, anddried under vaccum at 40 C to a constant
weight. Anequal sample was extracted by a modification of themethod
of Bligh and Dyer (2), and the lipid phos-phate was measured by the
method of Bartlett (1).
VOL. 109, 1972 669
-
RAY AND WHITE
was run in chloroform-methanol-diisobutyl-ketone-acetic
acid-water (12:5:23:13:2), and the seconddimension was run in
chloroform-methanol-diisobu-tylketone-pyridine-0.5 M ammonium
chloride, pH10.4 (15:9:13:18:3). Since
lysylphosphatidylglycerol(LPG) is liable in the second solvent, the
lower por-tion of the chromatogram was cut off as described byShort
and White (20) after the first solvent. A ra-dioautogram of the
lipid separation has been pub-lished (20).
Labeling of the phospholipids. Pulse-chase ex-periments were
done by growing 500 ml of cells to anabsorbance of 0.12 in
nonradioactive medium, andthen pulsing the cells for two
generations with 1.4mCi of HS32PO4. After this time, the cells were
har-vested by filtration, washed with an equal volume
ofnonradioactive medium, and resuspended in 500 mlof nonradioactive
medium with or without glycerol.Localization of the glycerol moiety
in this auxotrophwas achieved by growing the cells in medium
contain-ing 20 yg of glycerol per ml, supplementing with 100pCi of
glycerol-1,3-14C for 10 generations, and har-vesting the cells by
centrifugation. After washingtwice with 50 mm phosphate buffer (pH
7.4), sampleswere taken and counted and the lipids were isolatedas
described. After mild alkaline methanolysis (25),the fatty acids,
carotenoids, and vitamin K isopreno-logues could be separated from
the water-solubleglycerol phosphate esters. Incorporation
experi-ments were done as described in the text. The meas-urement
of the phospholipid com'position was doneby determining the
percentage of 32p in each phos-pholipid after growing the cells in
constant specific-activity medium for at least 10 generations.
Thepercentage of composition determined by the meas-urement of S2P
was consistent with the measurementof phosphate in at least two
experiments. Totalphosphate was determined by the method of
Bartlettafter digestion with 23% HCIO4 for 1 hr at 200 C, asadapted
for the autoanalyzer (1). Samples were as-sayed for radioactivity
on filter-paper discs in aPackard scintillation spectrophotometer
model 2311as described (10). A scintillation fluid of 9.28 mM2,
5-bis [2(5-terbutylbenzoazol) ]-thiophene in toluenewas used.
RESULTSThe glycerol auxotroph of S. aureus S-2 re-
quired glycerol at a concentration of 15 to 30 ,gper ml for
optimal growth. Below this con-centration the doubling time
increased, andbelow 4 ,ug per ml there was no visible growth.When
cells from the exponential phase ofgrowth were washed free of
glycerol and resus-pended in medium without glycerol, growthslowed
after 30 min (Fig. 2). The cessation ofgrowth, as indicated by
absorbance, occurredsimultaneously with the cessation of RNA
syn-thesis (Fig. 2). If, after depriving the cells ofglycerol for
90 min, glycerol was added, therewas a lag before the resumption of
RNA syn-thesis and growth. Protein synthesis was inhib-
ac0e ' BACTERIAL DENSITY,-0.5
z
0
I4II20 140 16020 40 60tONuw RNA SYNTHESIS
-J
a .
co
-J10
C.)
uje PROTEIN SYNTHESIS-I
w
20 40 60 80 100 I20 140 160MINUTES
FIG. 2. Effect of glycerol deprivation and readdi-tion on
ribonucleic acid synthesis and protein syn-thesis. Upper graph
shows the effect of glycerol dep-rivation on protein synthesis and
after the readditionof glycerol after starvation (marked by the
arrow).The cells were grown (for seven generations), har-vested,
washed, and resuspended in medium con-taining 1 uCi of serine-2- "C
per ml. At the timesindicated, 1-ml samples were withdrawn and
addedto 2 ml of cold 10% trichloroacetic acid, filtered onMillipore
filters, washed, and counted. Middle graphshows the effect of
glycerol deprivation on ribonu-cleic acid synthesis as determined
by growing thecells in "4C-uracil and measured as in the
middlegraph.ited after 30 min of glycerol deprivation; how-ever, it
required about 90 min for protein syn-thesis to cease. When
glycerol was added, evenafter 115 min of deprivation, protein
synthesisresumed at the predeprivation rates, after abrief lag.
Effect of growth phase on the phospho-lipid composition. The
phospholipid composi-tion of the glycerol mutant was somewhat
dif-ferent from that of the parent S. aureus U-71.In Fig. 3, the
phospholipid composition of S-2supplemented with glycerol was
compared at
670 J. BACTERIOL.
-
GLYCEROL AUXOTROPH OF S. AUREUS
different stages of growth. It can be seen thatduring the
exponential phase of growth thephospholipid composition was
constant: phos-phatidylglycerol (PG), lysylphosphatidylgly-cerol
(LPG), cardiolipin (CL), and phosphati-dic acid (PA) represented
92, 5, 1, and 2%of the total phospholipids, respectively. How-ever,
in early stationary phase, the proportionof PG decreased, the
proportions of LPG, CL,and phosphatidylglucose (PGlu) increased,
andthat of PA remained constant. In the parent,S. aureus U-71, PG,
LPG, and CL represent80, 12, and 5% of the total phospholipids,
re-spectively, during exponential growth (21). Themetabolism of the
phospholipids in the mutantwere examined in the exponential phase
ofgrowth, so that any changes in the metabolismoccurred under
conditions when the amount of
85
C*j 80In)
o75
J 70
o65
w
M 15
X. 10
1.0 .
EO.5
In.I-4 0.1.
0.050
uxm)4
5 6 7 8 9 10 11 12HOURS
FIG. 3. Effect of growth on the phospholipid com-position of
Staphylococcus aureus S-2. The cellswere grown in medium containing
32p at constantspecific activity for at least 10 generations. At
thepoints indicated, 25-ml samples were withdrawnfrom the flask and
added directly to 37.5 ml of chlo-roform and 75 ml of methanol as
described. The in-dividual phospholipids were separated by
chroma-tography on silica gel-impregnated paper, and theindividual
lipids were localized by radioautography.PG (0), LPG (A), CL (-),
PA (*), and PGlu (a).
each phospholipid remained constant for thosecells supplemented
with glycerol.
Localization of the glycerol incorporatedby the glycerol
auxotroph. When cells weresupplemented with "4C-glycerol, the
distribu-tion of the radioactivity showed that 41% ofthe total
"4C-glycerol incorporated into thecells was found in the lipid
fraction of the cells(Table 1). By mild alkaline methanolysis,
itwas shown that 96% of the total radioactivityin the lipid
fraction was found in the water-soluble glycerol phosphate
backbone. Only 4%of the labeled glycerol appeared in the
fattyacids, carotenoids, or vitamin K2 isopreno-logues.Turnover of
the phospholipids in cells in
the presence and absence of glycerol. Whencells grown in
glycerol were pulsed withH.32 P04 for two generations and washed
withmedium devoid of glycerol and 32P, there wasno lag in growth in
both cells supplementedwith glycerol and devoid of glycerol. In the
cul-tures without glycerol, it was seen that growthslowed after 36
min and completely ceasedafter 50 min (Fig. 4). Immediately after
thedeprivation of glycerol, net phospholipid syn-thesis ceased. The
turnover of the individualphospholipids was examined for two
genera-tions after the deprivation of glycerol. Themajor
phospholipid, PG, lost 55% of the 32P inone generation; and CL and
PA (not shown)lost 20% of the isotope in one generation inboth the
cells supplemented with glycerol and
TABLE 1. Localization of glycerol-i,3-' 4Cincorporated by
Staphylococcus aureus S-2a
Fraction Counts per min Per cent 14C
Whole cells ...... 2.05 x 107 100Lipid extract ..... 8.2 x 10'
41Fatty acidsb ...... 2.4 x 105 3Glycerol phosphate
esters ......... 7.8 x 10' 38
aCells were grown in 50 ml of medium containing20 ug of glycerol
and 2 gsCi of glycerol-1,3- 4C perml to an absorbance of 0.66 at
750 nm. The cellswere harvested by centrifugation and washed
twicewith 50 mm phosphate buffer (pH 7.4). Samples weretaken for
total radioactivity. The lipids were ex-tracted as described and
samples were counted onfilter-paper discs. Mild alkaline
methanolysis (25)was performed on the lipid extract, and the
fattyacids, carotenoids, and vitamin K2 isoprenologueswere
separated from the water-soluble glycerol phos-phate esters and
counted.bThe fatty acid fraction includes the carotenoids
and vitamin K2 isoprenologues; however, togetherthey represent
less than 1% of the fatty acids.
671VOL. 109, 1972
-
RAY AND WHITE
those without glycerol (Fig. 4). The major dif-ference in the
turnover rates of the plus andminus glycerol-grown cells was seen
in LPG.Without glycerol, LPG accumulated a twofoldincrease in
radioactivity. In the presence ofglycerol, LPG lost 25% of the
radioactivity
E
0
In
-I-4
i
z
44
0
0)o
after one generation.Effect of glycerol deprivation on the
phos-
pholipid composition. When cells were la-beled for 10
generations in medium contain-ing (constant specific activity)
H.32P0o, thepercentage of composition of the individual
15 30 45 60 75 90 105 120 135 150 165MINUTES
jI;
a. .060
a.
, .04
cnJ
0
A.02.
15
4co
w
lCi
.4c
I0-
0.
IC)
+GLYCEROL
%GLYCEROL
)5 120 135 150 165 180MINUTES
MINUTES
FIG. 4. Turnover of the phospholipids of Staphylococcus aureus
S-2, plus and minus glycerol. An expo-nentially growing culture of
S. aureus was pulsed for two generations with H,32PO4 (1.0 mCi/500
ml), filtered,washed with medium devoid of glycerol and
radioactivity, and resuspended in medium plus glycerol
(solidsymbols) and minus glycerol (open symbols). Samples were
withdrawn at the times indicated, and the lipidswere extracted and
separated as described. PG + glycerol (0), PG - glycerol (0), LPG +
glycerol (A), LPG- glycerol (A), CL + glycerol (U), and CL -
glycerol (0).
06 ' PULSE 32p0.50.4 1 +GLYCEROL
02.3 I(I -GLYCEROL
oi: .x .X 0-2 .I
672 J. BACTERIOL.
-
GLYCEROL AUXOTROPH OF S. AUREUS
phospholipids was determined by measuringthe radioactivity in
the individual phospho-lipids. The effect of glycerol deprivation
on thephospholipid composition of the mutant isseen in Fig. 5. The
top graph shows the effectof glycerol deprivation on (i) the total
phos-pholipid content and (ii) the total radioactivityin the
phospholipids. It can be seen by bothmeasurements that there is no
increase in thecontent of lipid phosphate after the depriva-tion of
glycerol. When glycerol was added tothe deprived cells after 60 min
of deprivation,
there was a 5-min lag before the content ofphospholipids started
to increase. Whenphospholipid synthesis resumed, the rate
ofsynthesis after the addition of glycerol wasthree times greater
than the growth rate. Thisrapid rate of synthesis continued for
about 25min. During the time of glycerol deprivation,when the total
phospholipid content remainedconstant, the proportions of the
individualphospholipids changed drastically. Immedi-ately after
glycerol deprivation, the PG de-creased from 92 to 45% of the total
lipid phos-
w
.10
..05z
..04 a-0
.503.02 o
rCEROL GLYCEROL / XXZ a <
20 40 60 80 100 120 140MINUTES
FIG. 5. Effect of glycerol deprivation on the phospholipid
composition. The cells were grown in mediumcontaining a constant
amount of H 32po4 to an optical density of 0.3, filtered, washed in
medium containingradioactivity and devoid of glycerol (as indicated
by the first arrow), and resuspended in medium containingH 32PO4
but devoid of glycerol. At the points indicated, samples were
withdrawn and added directly to 37.5ml of chloroform and 75 ml of
methanol. Top graph represents the total 32p and micromoles of
lipid phos-phate per sample in cells grown with glycerol (up to the
first arrow), after filtering, washing, and resus-pending in
deprivation medium (to the second arrow) and upon the readdition of
glycerol to the medium.Middle graph shows the growth of the culture
with glycerol, after filtering and depriving the cells of
glyceroland upon the readdition of glycerol. Bottom graph shows the
effect of glycerol deprivation on the phospho-lipid composition as
compared to cells grown in the presence of glycerol (O to 48 min)
and to prestarved cellsafter the addition of glycerol. PG (0), LPG
(A), CL (-), and PA (0).
673VOL. 109, 1972
-
RAY AND WHITE
phate. There was a concomitant increase inthe proportion of LPG
from 6.5 to 45% duringdeprivation. Upon the readdition of
glycerol,the proportions of PG started to increase im-mediately and
that of LPG decreased. Theproportions reached the predeprivation
levelsafter about 20 min. The proportion of CL re-mained relatively
constant both with and with-out glycerol. PA increased immediately
afterglycerol deprivation and amounted to 8% after7 min but then
decreased to the steady-statelevel even before the addition of
glycerol. Thisincrease in the proportion of LPG and the de-crease
in PG were not due to the lowering ofthe pH, since the pH remained
constant atbetween 7.2 and 7.1 during the experiments.
Incorporation of "4C-glycerol and 32P intothe phosphQlipids of
S. aureus. After 60 minof glycerol deprivation, glycerol-i, 3- 14C
wasadded, and the kinetics of the incorporation ofthe radioactivity
into the glycerol backbone ofthe lipids were measured (Fig. 6).
Radioactiveglycerol was incorporated most rapidly intoPA and PG.
The incorporation into LPG wassomewhat slower and that into CL was
evenslower. This incorporation was identical to theincorporation of
glycerol into cells under ordi-nary conditions before deprivation,
except thatthe rate of incorporation into PG was two tothree times
faster. The incorporation of thelabeled glycerol can be compared to
the incor-poration of 32P into phospholipids before glyc-erol
starvation, during starvation, and after 60min of glycerol
starvation followed by the ad-dition of glycerol (Fig. 7). Figure 7
shows thatthe radioactivity first appeared in PA, andthen PG, and
then LPG, and finally in CLunder normal growing conditions.
However,during glycerol starvation, when the phospho-lipid
composition of the cells has changed (seeFig. 5), the rate of
incorporation was six- tosevenfold lower than that of the
glycerol-sup-plemented cells. The radioactivity first ap-peared in
PA and LPG. There was a 5-min lagbefore radioactivity appeared in
PG (eventhough it is the precursor to LPG) and CL. Ifglycerol was
added after a 60-min period ofstarvation, the radioactive label was
first de-tected in PA followed by PG. There was a 5-min lag before
the incorporation of 32p intoLPG and CL.
Effect of glycerol deprivation on the fattyacids of S. aureus.
In S. aureus grown in me-dium containing constant specific-activity
"4C-acetate, the rate of incorporation of the acetatecan be used to
measure the amount of fattyacid synthesized, since (i) there was no
ra-dioactivity detected in the glycerol backbone,
-Ji PGw
Ij3
LPG
10
U
10 5 10 15 20 25 30MINUTES
FIG. 6. Incorporation of glycerol-1,3-14C inStaphylococcus
aureus after 45 min of glycerol depri-vation. The cells were
deprived of glycerol as de-scribed before; after 45 min, 20 gg of
glycerol wasadded per ml, containing 100 ;Ci of glycerol-1,3-14C,
and samples were taken at the times indicated.PG (-), PA (*), LPG
(A), CL (A).
and (ii) the amount of neutral lipid was negli-gible in relation
to the amount of fatty acids;the carotenoids equaled 0.1 ,mole and
the vi-tamin K2 isoprenologues equaled 0.2 gmoleper g of dry weight
versus 150 ,moles of fattyacids per g of dry weight. When S. aureus
wasdeprived of glycerol, fatty acid synthesisslowed almost
immediately to half the rate ofcells supplemented with glycerol
(Fig. 8). After75 min, glycerol was added and the rate imme-diately
increased. In normally growing cells ofS-2, the free fatty acid
content was about 1 to2% of the total extractable fatty acids.
Whenthe cells were deprived of glycerol, the contentof the free
fatty acids increased to 12.5% of thetotal fatty acids in an hour.
However, evenbefore the addition of glycerol, the contentdecreased.
With the addition of glycerol, thefree fatty acid content decreased
to the pre-deprivation level. The fatty acid compositionof this
mutant was altered by the deprivationof glycerol. Under conditions
where cells weresupplemented with glycerol the iso- and
anti-iso-branched-chain fatty acids of chain lengthC-14 to C-18
predominated. However, uponglycerol starvation, longer chain
length,branched fatty acids appeared (mainly iso-C-
674 J. BACTERIOL.
-
GLYCEROL AUXOTROPH OF S. AUREUS
a-cLJJ
2,L&J
cra.
CL
a.
l--)4
2
wa.
675
a.
w-J02
3 XC\Mn
2
La.
0
wa.I n
MINUTES MINUTES MINUTES
FIG. 7. Incorporation of 32p into the phospholipids of
Staphylococcus aureus S-2. The incorporation of 32pwas measured in
these lipids during normal phospholipid metabolism (left graph), in
the absence of glycerolafter 20 minutes of starvation, and after
the addition of 20 jAg of glycerol per ml to cells deprived of
glycerolfor 60 min. Arrow indicates the time of glycerol addition.
PG (0), LPG (A), CL (a), and PA (*).
21 and anti-iso-C-21, which represented 5 and10% of the total
fatty acids, respectively) andwere found in the free fatty acids.
After 60min, the longer chain fatty acids were esteri-fied into the
phospholipids. The longer chainfatty acids were still present 30
min after theaddition of glycerol to glycerol-starved cells.
DISCUSSIONExponentially growing glycerol auxotrophs
of S. aureus, when deprived of glycerol, showedan abrupt halt in
the net increase in mem-brane phospholipids even though growth,RNA,
and protein synthesis continued for 30to 90 min (Fig. 2 and 5).
During this period,there was an increase in the content of LPGand a
concomitant decrease in PG (Fig. 5);however, this was not simply a
conversion ofold PG to LPG since 32p was incorporatedquite rapidly
into LPG (Fig. 7). The PG inglycerol-deprived cells lost 32P at the
samerate as cells grown with glycerol and synthe-sized PG very
slowly so that the fall in theproportion of PG was balanced by the
increasein LPG (Fig. 3 and 4). In the period of
glyceroldeprivation, the total lipid phosphate and theproportion of
CL remained constant (Fig. 5).Since the pathway of LPG synthesis in
S. au-reus involves PA-> PG-> LPG (4, 15), it isclear that
during glycerol deprivation in thismutant (i) there was a rapid
synthesis of LPG
that involved a portion of the PG pool with aspecific activity
45 times greater than that ofthe rest of the PG pool (Fig. 7) and
(ii) therewas an accumulation of 32p in LPG for 20 minfollowed by
no turnover in the absence of glyc-erol, in contrast to turnover of
LPG duringgrowth with glycerol (Fig. 4). This representsanother
example of the heterogeneity in themetabolism of the membrane
lipids. Differ-ences in the metabolism or portions of the PGand CL
pools have been demonstrated in S.aureus (20) and Haemophilus
parainfluenzae(22, 25). During glycerol deprivation, thestriking
features of phospholipid metabolismwere the rapid synthesis of LPG
and the cessa-tion of its normally slow turnover. Previouswork with
both B. subtilis (16) and S. aureus(4, 6) suggested that the
metabolism of LPGresponded to the pH of the growth medium.During
the accumulation of LPG in glyceroldeprivation, the pH remained
constant.The glycerol auxotroph not only accumulates
LPG and slows incorporation of 32p into thelipids when deprived
of glycerol but whengrown with glycerol it showed differences
fromthe parental type. The mutant grown withglycerol has roughly
half the total phospholipidper g dry weight (30 to 35 umoles) as
the wildtype (60 to 65 jimoles/g dry weight, reference21). The
mutant forms much less total CL, andthe CL has a much slower
metabolism thanthe wild type (Fig. 4, reference 20). During the
VOL. 109, 1972
-
RAY AND WHITE J. BACTERIOL.
E0
W 0.6 CONTROLuj 0.4 +GLYCEROL
0.20
co
-
GLYCEROL AUXOTROPH OF S. AUREUS
synthesized fatty acids has been postulated toexplain the fatty
acid differences in the phos-pholipids of S. aureus (24), and the
cycle hasbeen demonstrated in H. parainfluenzae whenthe
wall-membrane complex is damaged (22).When glycerol was added to
the deprived
culture, both glycerol and 32P were rapidlyincorporated into the
lipids (Fig. 6 and 7). Thisoccurred in the presence of 0.1 mm
chloram-phenicol, which suggested that the
phospho-lipid-synthesizing enzymes were presentduring the
deprivation. The major features ofthe recovery were the rapid
synthesis of PGand the slower synthesis of LPG.A shift in lipid
metabolism was also de-
tected in a glycerol-requiring mutant of B.subtilis (10). In
this mutant, the free fatty acidaccumulation was less pronounced
and thelipid that accumulated during deprivation wasCL. In both
these glycerol auxotrophs, glyceroldeprivation stopped net
phospholipid syn-thesis and shifted the proportions of the
phos-pholipids, but allowed some de novo lipid syn-thesis and
catabolism. Perhaps this limitedmetabolism explains the normal
induction ofthe lactose (14) and citrate operons (26) butthe poor
function of these membrane systems.
ACKNOWLEDGMENTS
We thank L. Mindich for sending us the glycerol auxo-troph, S.
aureus S-2. This investigation was supported byPublic Health
Service grant 1-FO2-GM 45691-01 from theNational Institute of
General Medical Sciences to P. H.Ray, grant GM-10285 from the
National Institute of GeneralMedical Sciences, and Public Health
Service grant GB-17984 from the Metabolic Biology Section of the
NationalScience Foundation to D. C. White.
LITERATURE CITED
1. Bartlett, G. R. 1959. Phosphorous assay in column
chro-matography. J. Biol. Chem. 234:466-468.
2. Bligh, E. G., and W. J. Dyer. 1959. A rapid method oftotal
lipid extraction and purification. Can. J.Biochem. Physiol.
37:911-917.
3. Fox, C. F. 1969. A lipid requirement for inducation oflactose
transport in Escherichia coli. Biochemistry 63:850-855.
4. Gould, R. M., and W. J. Lennarz. 1970. Metabolism
ofphosphatidylglycerol and lysyl phosphatidylglycerolin
Staphylococcus aureus. J. Bacteriol. 104:1135-1144.
5. Henning, U., G. Dennert, K. Rehn, and Gisela Deppe.1969.
Effects of oleate starvation in a fatty acid auxo-troph of
Escherichia coli K-12. J. Bacteriol. 98:784-796.
6. Houtsmuller, U. M. T., and L. L. M. Van Deenen. 1965.On the
amino acid esters of phosphatidyl glycerolfrom bacteria. Biochim.
Biophys. Acta 106:564-576.
7. Hsu, C. C. and C. Fred Fox. 1970. Induction of the lac-tose
transport system in a lipid-synthesis-defectivemutant of
Escherichia coli. J. Bacteriol. 103:410-416.
8. Kundig, W., and S. Roseman. 1971. Sugar transport.
II.Characterization of constitutive membrane-boundenzyme fl of the
Escherichia coli phosphotransferasesystem. J. Biol. Chem.
246:1407-1418.
9. Joyce, G. H., R. K. Hammond, and D. C. White. 1970.Changes in
membrane lipid composition in exponen-tially growing Staphylococcus
aureus during the shiftfrom 37 to 25 C. J. Bacteriol.
104:323-330.
10. Lillich, T. T., and D. C. White. 1971. Phospholipid
me-tabolism in the absence of net phospholipid synthesisin a
glycerol-requiring mutant of Bacillus subtilis. J.Bacteriol.
107:790-797.
11. Milner, L. S., and H. R. Kaback. 1970. The role of
phos-phatidylglycerol in the vectorial phosphorylation ofsugars by
isolated bacterial membrane preparations.Proc. Nat. Acad. Sci.
U.S.A. 65:683-690.
12. Mindich, L. 1970. Membrane synthesis in Bacillus sub-tilis.
I. Isolation and properties of strains bearingmutations in glycerol
metabolism. J. Mol. Biol. 49:415-432.
13. Mindich, L. 1970. Membrane synthesis in Bacillus sub-tilis.
II. Integration of membrane proteins in the ab-sence of lipid
synthesis. J. Mol. Biol. 49:433-439.
14. Mindich, L. 1971. Induction of Staphylococcus aureuslactose
permease in the absence of glycerolipid syn-thesis. Proc. Nat.
Acad. Sci. U.S.A. 68:420-424.
15. Nesbitt, J. A., III, and W. J. Lennarz. 1968. Participa-tion
of aminoacyl transfer ribonucleic acid in
aminacylphosphatidylglycerol synthesis. I. Specificity of
lysylphosphaticylglycerol synthetase. J. Biol. Chem.
243:3088-3095.
16. Op den Kamp, J. A. F., I. Redai, and L. L. M. van Dee-nen.
1969. Phospholipid composition of Bacillus sub-tilis. J. Bacteriol.
99:298-363.
17. Ray, P. H., D. C. White, and T. D. Brock. 1971. Effectof
temperature on the fatty acid composition ofThermus aquaticus. J.
Bacteriol. 106:25-30.
18. Rothfield, L., and A. Finkelstein. 1968. Membrane
bio-chemistry. Annu. Rev. Biochem. 37:463-496.
19. Schairer, H. U., and P. Overath. 1969. Lipids
containingtransunsaturated fatty acids change the
temperaturecharacteristic of thiomethylgalactoside accumulationin
Escherichia coli. J. Mol. Biol. 44:209-214.
20. Short, S. A., and D. C. White. 1970. Metabolism of
theglucosyl diglycerides and phosphatidylglucose ofStaphylococcus
aureus. J. Bacteriol. 104:120-132.
21. Short, S. A., and D. C. White. 1971. Metabolism
ofphosphatidylglycerol, lysylphosphatidylglycerol, andcardiolipin
of Staphylococcus aureus. J. Bacteriol.108:219-226.
22. Tucker, A. N., and D. C. White. 1971. Detection of arapidly
metabolizing portion of the membrane cardi-olipin in Haemophilus
parainfluenzae. J. Bacteriol.108:1058-1064.
23. White, D. C., and R. H. Cox. 1967. Identification
andlocalization of the fatty acids in Haemophilus parain-fluenzae.
J. Bacteriol. 93:1079-1088.
24. White, D. C., and F. E. Frerman. 1967. Extraction,
char-acterization, and cellular localization of the lipids
ofStaphylococcus aureus. J. Bacteriol. 94:1854-1867.
25. White, D. C., and A. N. Tucker. 1969. Phospholipidmetabolism
during bacterial growth. J. Lipid Res. 10:220-233.
26. Willecke, K., and L. Mindich. 1971. Induction of
citratetransport in Bacillus subtilis during the absence
ofphospholipid synthesis. J. Bacteriol. 106:514-518.
27. Wuthier, R. E. 1966. Two-dimensional chromatographyon silica
gel-loaded paper for the microanalysis ofpolar lipids. J. Lipid
Res. 7:544-550.
677VOL. 109, 1972
